Re-appearing barlines
I've encountered a strange and frustrating problem in notating a solo cadenza in a J.G. Graun concerto.
Tradionally, as most of you probably know, cadenzas are written in unmetered, un-barred notation. If this were a bit shorter, I would have just made one long, irregular measure, but it's actually 60 beats long, and a few months ago when I wrote this section, the program started arguing with me about the number of beats I was asking for. When I tried to paste it into a specially-designated 60-beat-long measure, the program crashed and upon reopening I was told the file was corrupted. I had to scrap that file and start over from the back-up copy. I obviously didn't try that again. In fact, I was too tired to argue, so I just let it break into 5 measures of 12/8 and figured I'd make all the rests and barlines invisible and remove each measure from the measure count individually.
That's what I did...but today when I started proofing the score for print I noticed the barlines I'd made invisible earlier had all changed back to visible. Not the rests, just the barlines. Oh, well, I shrugged; a PITA, but I did it over again. Then, as I was clicking on the final group of barlines I wanted invisible, I SAW the already-invisible barlines jump back to visibility.
I tried a few alternatives, including setting each barline invisible independently (didn't make any difference) or setting them ALL invisible in one group (that didn't work either; they all went invisible when I unchecked the 'visible' box in the inspector, then, at the next click I made elsewhere on the score, they all jumped back to visible. I also thought of setting the barline to print 'white' (but that leaves gaps in the staffs) and even using 'dotted' barlines (which I thought I could tolerate), but those popped back to standard, solid, barlines just the way the invisible ones became visible again. It's driving me nuts, as you can imagine.
It appears to be happening only in the score; I THINK I have managed to make the barlines stay invisible in the parts.
Is this a known issue, and if so, has a work-around been created? (And if not, can someone come up with one?) Or have I made an error in the way I'm doing this (using the inspector, that is)?
Comments
Well, first of all there are better ways to get the effect you're going for—the simplest is to select all the measures and go up to Edit -> Measure -> Join Measures. If for whatever reason you want to keep separate measures with invisible barlines, the simpler way to handle that would be to secondary-click on just one barline, choose Select -> All Similar Elements from the context menu, and press [v] to toggle visibility. But things made invisible should definitely stay invisible, whatever way you do it. Can you attach the score here with step-by-step instructions to reproduce the issue?
In reply to Well, first of all there are by Isaac Weiss
Join measures? Just when I thought I knew most of the features! Okay, I'll try that and see if it causes a crash. I'll save a copy first.
I thought of using 'select all similar elements' but wouldn't that select ALL the barlines in the entire score? Is there a way to limit that selection process to a specific set of measures...?
Toggle (v)--check. That will save me a lot of mousing around in future. Thx.
As for posting the score, I can't post the whole thing--it's very long--but I'll try to make a cut-and-paste of the section showing the problem, hoping that whatever's causing the problem will come along for the ride so you can see it.
Back in a while.
In reply to Join measures? Just when I by Recorder485
Length shouldn't preclude posting the score.
Join measures is new for 2.0, as are tons of other features, so maybe that's why you didn't know about it.
You can limit any selection to a range by selecting the range first then roght clicking an element and choosing Select / All similar elements in range selection (also new for 2.0).
In reply to Length shouldn't preclude by Marc Sabatella
The other reason I can't post the whole score is that it's a transcription about to be published commercially and the publisher has an ironclad rule against posting any digital file on an open website. I could only post this to a private group on my Musescore user page, but you'd have to write me by private e-mail to get the log-in codes for that.
I will re-try the trick of selecting the group of measures before hitting 'select all similar elements' but when I've tried that in the past, the proper menu items don't show up on right-click.
In reply to The other reason I can't post by Recorder485
Not just "all similar elements", but "all similar elements in range selection".
Preumably you could delete the actual notes (Ctrl+A, Delete) leaving just the empty measures and post that.
In reply to Not just "all similar by Marc Sabatella
Sorry, I missed this message amongst the back-and-forth of all the others. You're saying that if I post the score with all the music deleted you can still see any underlying code problem but there would be no way for anyone to 'undo' the delete of the music itself once they had the file? If that's the case, I could take the original file and do as you suggested.
In reply to Sorry, I missed this message by Recorder485
Correct, undo works only during the current editing session. No record of what was deleted would be aved to the file.
In reply to Correct, undo works only by Marc Sabatella
Okay, give me a few minutes and I'll post a copy of the blanked master file. As I have a clean working copy of it under a different name, it shouldn't make any difference if I save a blanked score to the original file name.
In reply to Okay, give me a few minutes by Recorder485
Here it is. The original problem occured at the end of the second movement, roughly in the vicinity of measure 202 et seq. Note that when I blanked the file, a lot of the special formatting in that area went away, too (for instance, the rests didn't argue about staying invisible, but they are visible now). I hope this helps.
In reply to Here it is. The original by Recorder485
Are there particular steps you could give me to demonstrate the problem? I tried a few random things and didn't notice any problems.
In reply to Are there particular steps by Marc Sabatella
I'm stumped. I went back to the original file on my computer--the one I blanked out to post here--and the problem is no longer in that file, either. Either it was embedded in the music itself--which makes no sense, because a COPY of that file WITH the unaltered music, was saved under a new name and the problem did not/does not appear in that file--or it was a 'ghost in the machine' and has flown off with George and Marion Kerby to replenish its ectoplasm in a speakeasy somewhere....
Sorry for the false alarm, but I swear, I battled unsuccessfully with that 'ghost' for two hours before posting my question.
In reply to Length shouldn't preclude by Marc Sabatella
You were right--I wasn't right clicking on a specific element, just on the whole element group. Right clicking on one of the barlines brought up the proper menu, and using 'select all similar items in range' worked perfectly. Thanks.
In reply to Well, first of all there are by Isaac Weiss
Well, I must report that (a) the original problem did NOT carry over to the sample score I created, so I can't reproduce it for you. In that paste-in, I can (and did, using your neat toggle-v trick, blank out all those barlines quickly and without fuss. So whatever code hiccup is in my original file stayed behind when I copied those measures to the clipboard.
(b) The idea of using 'join measures' appealed to me quite a lot, but the result (performed on the sample score spoken of above) was not satisfactory. Take a look at the screen-shot, and let me know if there's a way to avoid this. Heh....
In reply to Well, I must report that (a) by Recorder485
What about that are you not liking? The long rests? You could just enter different durations if you like, or mark them invisible. If you mean the fact that the notes don't physically fit on one line, well, you just need smaller notation - either reduce the staff size or the note size or note spacing or some combination of these.
In reply to What about that are you not by Marc Sabatella
I am puzzled; it seems you and I aren't seeing the same thing. What I'm seeing is notes that run right off the staves to the right, into the margin and beyond, and disappearing.
If you're seeing that, then I'm even more puzzled. Changing the scaling or note size or whatever just to force that section to fit on one line is not a viable typographic option. It would be equivalent to a newspaper making the type smaller on one line in order to fit one long, unhyphenatable word into the column.
I don't suppose there's a way to 'force' Musescore to break a line in the middle of a measure? That would actually be quite useful if one were trying to create a typset facsimile of an old MS, where the copyists often broke measures into two pieces.
In reply to I am puzzled; it seems you by Recorder485
That's what I meant by "the notes don't physically fit on one line". You can't expect to fit infinitely many notes on one line of music, after all. You could elect to split that into two measures with part of the cadenza on one line, part on the next - use the Split command, right next to the Join command. Or just don't join it all into one measure in the first palce but instead choose a sensible break point and only join up to there.
Or, as I said, just make everything smaller - which *is* exactly how many publisher print cadenzas and similar passages. I have much music published that way.
How did you do it in the version using invisible barlines? It be even more impossible to fit it all one one line that way.
In reply to That's what I meant by "the by Marc Sabatella
Hah! Another feature I hadn't noticed. I will try 'split measure', just to see how it works.
I do know that music publishers often (almost always, I would say) print cadenzas in small notation or on small staves. And yes, obviously, I can't expect to fit an infinite number of notes onto one line of music. But there's a limit to how much shrinking can be gotten away with before it starts to look like one of Pisendel's manuscripts. ;o)
This is somewhat akin to the 'cue' function we discussed a while back; it would be really convenient from a user's point of view to have some sort of dedicated cadenza function that would allow us to set a cadenza in a meterless, unbarred format without having to manually fiddle around with line breaks, setting note size to 'small', exempting measures from the measure count, and making rests in the non-playing parts of the score invisible. I guess that from a programmer's point of view, those demands would be a major pain--they require stepping outside a long list of established parameters--but, would it be possible to design something like that?
In reply to Hah! Another feature I hadn't by Recorder485
Seems the sort of highly specialized applications best suited to a plugin. But really, doing the things you mention takes only seconds - surely only a small fraction of the time it takes to actually enter the cadenza.
In reply to Seems the sort of highly by Marc Sabatella
Heh. It only takes seconds if nothing goes blooey in the process, which is what happened to me in this case. ;o)
In reply to Heh. It only takes seconds if by Recorder485
Yes, but again, that's only because you tried to put more stuff on one line than fits. If you had only tried to join as many notes as you were fitting on one line before, it would have worked perfectly. Or, do what i suggested - first join everything, see how much fits, then decide for yourself where a good place to split the line would be, and use the split fucntion. It's a subjective decision where to make the line break - it's not just about how many notes fit, but based on your own opinion of where the msot "logical" break would be. Thast's why it's best left to the user, and again, can be done in seconds using join & split.
In reply to Yes, but again, that's only by Marc Sabatella
I understood what you said earlier in that context and I don't disagree; but my previous comment about things going 'blooey' was made in the ORIGINAL context of having those barlines keep coming back like an unpaid bill.
I don't suppose we'll ever find out what happened in that file; the original is now gone-gone and only the blanked-out version and (thank Bog!) the problem-free copy remain. All I can say now is thanks for wanting to help. That means a lot, and the willingness of the MuseScore development team to address problems RIGHT NOW is probably the single biggest reason I won't even consider switching to a commercially-produced scorewriter like Finale or Sibelius. I simply can't imagine getting anything NEAR that kind of commitment out of the 'complaint management' (as opposed to 'complaint resolution') department of a major commercial software vendor.
You guys are great. Thanks again.
In reply to I understood what you said by Recorder485
You are welcome, and I completely agree with the sentiment!
Well, my immediate problem seems to be solved. By working from the backup copy I created of this score, and using Zack's 'toggle-v' technique combined with Marc's 'select all similar elements in range' technique, I seem to have been able to blank out those unwanted barlines. So--unless it jumps back out at me later--I'm good with the workaround.
I am sorry I wasn't able to post the score with the code problems in it; if anyone really wants to see it, please write me by private e-mail and I'll make arrangements.
In reply to Well, my immediate problem by Recorder485
Really, [v] isn't a separate technique—it just controls the exact same thing as the checkbox in the Inspector. But I'm glad you got something that works!
In reply to Really, [v] isn't a separate by Isaac Weiss
Keyboard shortcuts are the best thing since sliced bread and folding umbrellas!